The proposed secondary analysis seeks to discover the positive effects of past and recent volunteer work on health outcomes among participants in the Wisconsin LongitudinalStudy (WLS), using both core and sibling samples. Past research has shown that volunteer work has positive effects on both mental and physical health - and even mortality - in middle aged and elderly populations. Largely lacking in the research on this topicto date have been: 1) investigationof the mechanisms by which volunteering has its effects, 2) consideration of contextual effects and possible interactions with individual level factors, and 3) study of the effects of different kinds and amounts of volunteering. This research attempts to fill these gaps usingappropriate but previouslyunderused data from the WLS. Health outcomes to be examined include psychological well-being, depression, self-reported health, number and type of illnesses, and mortality. Potential moderators of the volunteering-health relationship include age, education, Lncome,physical activity,marital and work status, and social support. Psychological well-being and depression are expected to mediate the impact of volunteering on physicalhealth outcomes. The research will also investigate the effects of "intensity"and "diversity" of volunteer experience, and differential impacts of different kinds of volunteering. Finally, the research will explore the interactionsof contextualvariables reflecting social disadvantage at the county and census tract levels with individual level social participation variables in influencing health outcomes, which has not been done before. The proposal represents a significant new direction in the work of the principal investigator. Although her recent work on blood donation and volunteering involvedsurveys, she has not worked with the WLS sample or focused on health outcomes. The social participationvariables in the WLS have not received much attention, and considerable work needs to be done to refine them. Since further moderating and mediating variable measures will be added in the 2002-2003 wave of the WLS, and the social participation variables will be repeated, methodologicalwork on these measures is timely.